In close concert with projects 1 and 2, we seek common molecular features of abnormal bone collagen expression in tissue from mice and human cases of osteogenesis imperfecta (01) caused by mutations in genes encoding the CRTAP/P3H1/CYPB and FKBP10/HSP47 complexes and newly identified genes causing recessive 01. The significance and inter-relationships of defective prolyl 3-hydroxylation, associated post-translational over modification of lysine residues and abnormal cross-linking are the focus and basis of hypothesis-driven studies. One goal is to thoroughly test the possibility that abnormal post-translational chemistry, and in particular cross-linking, underlies the brittle bone phenotype. Though the focus is novel recessive forms of 01, the significance of the findings is likely to extend across all forms of 01. The specific aims are directed at establishing the cross-linking phenotype of bone collagen from mouse models and available human 01 cases, seeking a common pathology and common underlying mechanism.